Distribution of the CCR5delta32 allele (gene variant CCR5) in Rondônia, Western Amazonian region, Brazil.

Since around 1723, on the occasion of its initial colonization by Europeans, Rondonia has received successive waves of immigrants. This has been further swelled by individuals from northeastern Brazil, who began entering at the beginning of the twentieth century. The ethnic composition varies across the state according to the various sites of settlement of each wave of immigrants. We analyzed the frequency of the CCR5Δ32 allele of the CCR5 chemokine receptor, which is considered a Caucasian marker, in five sample sets from the population. Four were collected in Porto Velho, the state capital and the site of several waves of migration. Of these, two, from the Hospital de Base were comprised of HB Mothers and HB Newborns presenting allele frequencies of 3.5% and 3.1%, respectively, a third from the peri-urban neighborhoods of Candelária/Bate-Estaca (1.8%), whereas a fourth, from the Research Center on Tropical Medicine/CEPEM (0.6%), was composed of malaria patients under treament. The fifth sample (3.4%) came from the inland Quilombola village of Pedras Negras. Two homozygous individuals (CCR5Δ32/CCR5Δ32) were detected among the HB Mother samples. The frequency of this allele was heterogeneous and higher where the European inflow was more pronounced. The presence of the allele in Pedras Negras revealed European miscegenation in a community largely comprising Quilombolas.


Introduction
In Brazil, the contribution of genes of Amerindian, African and European origin to the population has assumed various different patterns in the Amazonian Region, to the point that, in the State of Rondônia, migration became a significant factor in the diversification of ethnic composition. The initial colonization through a succession of emigrational waves was maintained, during the 1970's, when the capital, Porto Velho, and the nearby regions received numerous workers, natives of the southern and southeastern regions, as well as individuals descended from Italians, Germans and other nationalities, this intense gene flow giving rise to a tri-ethnic population pattern (Ferreira et al., 2002).
The CCR5D32 allele appears to have originated quite recently (approximately 7,000 years ago) in northeastern Europe (Lidén et al., 2006). Although its frequency has now reached a relatively high level in Europeans,e.g.,16.3% in Finns and 15.8% in Moravians, it is not present among African populations, and is only so at low levels in the Asian (Martinson et al., 1997). Hence, by application it is possible to evaluate the influence of the European population on the genetic constitution of others.
According to Passos and Picanço (1998), in Brazil, its distribution varies in accordance to geographical location and ethnicity, i.e., 3.5% in the urban population of southeastern Brazil and 0% in Amerindians. Imbalances in the geographical distribution of the CCR5D32 allele might either enable certain populations to show better adaptive responses to a series of environmental threats (Galvani and Novembre, 2005) or, on the contrary, be a risk factor in the development of a number of infections (Vargas et al., 2009). According to Faure and Royer-Carenzi (2008), the distribution of this allele would depend on various genetic and/or environmental factors, such as the pathogen nature, thus resulting in an advantageous or disadvantageous relationship to the host.
Other genetic polymorphisms that occur in the CCR5 gene have also been described, viz., A29S, L55Q and R60S, which may be associated with HIV-1 infection (Howard et al., 1999;Teixeira et al., 2009;Hütter and Ganepola, 2011). According to Sullivan et al. (2001), the presence of the CCR5D32 allele in a population could possibly influence the epidemiological trends of HIV infection, since the prevalence of HIV/AIDS is higher in populations that present low CCR5D32 allele frequency compared with those with a high frequency of homozygous or heterozygous individuals.
According to data from the Health Secretariat of Rondônia, there was a rise of 11.8% in the incidence of AIDS between 1997 (5.8%) and 2009 (17.6%), given a margin of 100,000 inhabitants. Consequently, the objective here was to determine CCR5D32 allele frequency in certain localities with different settlement backgrounds. The presence of Caucasian genes was predicted to reflect the gene flow coming from historical migration waves.

Populations and areas of study
The study was carried out with samples collected in two regions of Rondônia, the city of Porto Velho (63°54'14" W; 08°45'43" S) and the rural village of Pedras Negras (62°54' O; 12°51' S). The first three were from Porto Velho, one from the city's riverine region and the remainder from other locations. The fourth subsample was collected in the rural region of the state in the village of Pedras Negras, situated in the Guapore River valley. A map of the region is presented in Figure 1.

Blood collection
After obtaining written consent, 5 mL of peripheral blood were collected in EDTA vacutainer tubes. In the case of newborns from the Base Hospital, blood was collected from the umbilical cord, transported in refrigerated containers to the Laboratory of Human Genetics of the IPEPATRO, and processed within 10 h of leukocyte collection. All the samples were stored in aliquots at -20°Cuntil further use.

DNA extraction
Genomic DNA was extracted from leukocytes using 300 mL of peripheral blood according to the protocol described by Higuchi (1989).
Detection of the CCR5 polymorphism CCR5 genotyping was performed by PCR using specific primers (CCR5Forward: 5' -GGTCTTCATTACAC CTGC -3' and CCR5Reverse: 5' -AGGATTCCCGAG TAGCAGATG -3') and the protocol described by Chies and Hutz (2003). PCR reactions were prepared in a final volume of 25 mL as follows: 3 mL of genomic DNA, 2.5 mL of 10X PCR buffer (200 mM Tris-HCl, pH 8.4, 500 mM KCl), 2.5 mL of a 200 mM dNTP mix, 1.5 mL of 50 mM MgCl 2 , 0.14 mL of each primer at 0.25 mM and 0.15 mL of 28 Farias et al. Taq DNA polymerase (5 U/mL; Invitrogen Corporation, San Diego, CA, USA). The conditions for thermocycling were: 1 cycle of 5 min at 94°C, 30 cycles of 1 min at 94°C, 1 min at 63°C and 1 min at 72°C, followed by a 10 min final extension at 72°C (Eppendorf Mastercycler gradient, Eppendorf AG, Hamburg, Germany). The PCR amplification products were subjected to electrophoresis on a 10% polyacrylamide gel (100 V, 1 h 30') and stained with 10% silver nitrate. The CCR5/CCR5 genotype was detected as a single band of 137 base pairs (bp), the CCR5/CCR5D32 heterozygous genotype as a band of 137 bp plus another of 105 bp, and the CCR5D32/CCR5D32 homozygous genotype as a single band of 105 bp.

Statistical analysis
Statistical analysis was performed using the GENEPOP 4.0 (Rousset, 2008) and SPSS 19.0 Programs. Allele frequencies were obtained by maximum likelihood, and adherence to Hardy-Weinberg equilibrium tested by the chi-square test. 95% confidence intervals were calculated. A first attempt to estimate the degree of racial mixing in the present study-population, undertaken with only two of the sample groups, viz., parturient women (HB Mothers) and newborns (HB Newborns), was obtained using the ADMIX95 software developed by Chakraborty (1985). The genetic systems, previously validated in the same two sample groups, and their associated parental frequencies, were CCR5, CYP2E1 (cytochrome P450, subfamily IIE, polypeptide 1), GSTP1 (glutathione S-transferase class Pi) and ACP1 (acid phosphatase of low molecular weight)

Ethic procedures
This research was approved by the Brazilian Council of Research Ethics (CONEP, Number 13356).

Results and Discussion
The distribution frequencies of the CCR5D32 allele in the samples from Porto Velho and Pedras Negras, two towns located in the State of Rondônia (Table 1), were heterogeneous and reflected the settlement history of Rondô-nia (Pinto, 1993), with higher frequencies in locales where the European flow was more intense (Teixeira and Ribeiro da Fonseca, 1998).
Ethnicity in the two samples from the Hospital de Base was predominantly African (64 and 65% for the HB Mothers and HB Newborns, respectively). In these samples, the D32 allele frequency was 0.035 ± 0.153 (p = 0.0477; c 2 = 7.136) among mothers and 0.031 ± 0.030 (p > 0.05; c 2 = 0.315) among newborns. Although the European contribution was higher in the filial than the maternal generation (17% vs. 5.9% Caucasian genotypes, respectively), two women homozygous for D32 were detected among the mothers. The first Brazilian report of individuals homozygous for the CCR5D32 allele was provided by Grimaldi et al. (2002), who, while investigating the occurrence of this allele in various ethnic groups from various regions, observed that its frequency varied according to ethnicity. Anthropogenetic data, collected in the interviews prior to blood collection, indicated that the HB sample was composed of individuals, who were either natives of other regions of the country or of other northern states (24.5% of the sample). Approximately 7.94% of the individuals in the study were from the northeastern region. Contributions from the other regions (midwestern, southern and southeastern) were lower.
The lowest CCR5D32 allele frequency of 0.0006 ± 0.000 (p > 0.05; c 2 = 0.003) was observed in the Porto Velho CEPEM subsample, with only one individual bearer. The Porto Velho riverine subsample (represented in this study by CBE) was comprised of individuals from the first emigration movements, mainly Barbadians and individuals from the northeast. While CCR5D32 allele frequency in this population was 0.018 ± 0.013 (p > 0.05; c 2 = 0.037), the allele itself was only observed in the heterozygous state. This low frequency could be expected, given the background of settlement by Negroid populations and miscegenation with the Amerindian people, present during the first colonization movements along the Madeira River (Ferreira, 1987). Pedras Negras, located in the southern CCR5 allele in Rondônia, Brazil 29 Guapore Valley (midwestern Rondônia), was originally populated by the descendants of slaves who had escaped from gold mining operations in the former capital of the State of Mato Grosso, between 1734 and 1835, thus having arrived in the Guapore Valley at least 200 years ago. This is the only region in Rondônia with a population that was originally Quilombolan (Teixeira MAD, 2004, PhD Thesis, Universidade Federal do Pará, Belém). The current inhabitants are mostly descendants of the union between Negroids and Caucasians, although the population also features a maternal native-Amerindian heritage. The observed frequency of the D32 allele in this group was 0.034 ± 0.018 (p > 0.05; c 2 = 0.037). Allele frequencies, similar to those found in the Rondonian samples, have been detected in other Brazilian regions, i.e., 0.6% in Cepem, 0.7% in African-Brazilian populations from Pará (Carvalhaes et al., 2004), 0.9 and 1.0% in Bahia (Carvalho et al., 2004), 1.8% in Candelária and Bate-Estaca and 1.9% and 2.0% in African-Brazilian individuals in Rio de Janeiro (Chies and Hutz, 2003) and Paraná (Boldt et al., 2009), respectively. Among HIV + subjects, frequencies ranged from 2.1% in Rio de Janeiro (Teixeira et al., 2009) to 2.4% in Rio Grande do Sul (Vieira et al., 2011), 2.6% in a miscegenated population in Bahia (Grimaldi et al., 2002) and 3.8% in São Paulo (Rigato et al., 2008). Hünemeier et al. (2005) found frequencies of 2.0 and 3.0% in Amerindian populations from south Brazil. Boldt et al. (2009) observed frequencies of at least 0.9% in Amerindians from the state of Paraná. The presence of the D32 allele in Amerindian populations is an indication of historical European miscegenation. The frequencies detected in our samples (3.1% in the HB Newborns, 3.4% among the Quilombolas of Pedras Negras and 3.5% of the HB Mothers sample) were similar to those found in an urban population of Belém (3.0%; Carvalhaes et al., 2004), an urban population of Ribeirão Preto, São Paulo (3.5%) and a mixed population in Alegrete, RS (3.8%). However, our frequencies were lower than the 4.4% observed in the Euro-Brazilian population of Porto Alegre, and the 4.7% and 5.8% of patients with sickle-cell anemia, thus of African descent, from Recife and Porto Alegre (RS), respectively (Chies and Hutz, 2003). Brazilian populations described as Afro-Brazilian presented allele frequencies of 5.6% in the state of Bahia (Carvalho et al., 2004) and 8.0% in the state of Rio Grande do Sul. The populations described as Euro-Brazilians generally presented the highest frequencies in the southern states: 6.5% in Joinville, State of Santa Catarina; 6.5% in Alegrete, State of Rio Grande do Sul; and 9.3% in the State of Paraná (Grimaldi et al., 2002;Vargas et al., 2006;Boldt et al., 2009, respectively). Sullivan et al., (2001) noted that the heterozygous presence of the mutant CCR5 allele can influence HIV epidemiological trends within a population, possibly through limiting the epidemic by reducing the likelihood of viral transmission by infected heterozygous individuals. In 2009, out of 150 HIV tests administered in the Vale do Guaporé, 80 proved to be positive. According to the state's health department, the northern region of Brazil (which includes Rondônia) exhibited a 39.6% increase in AIDS cases from 2002 to 2006. This high regional incidence incited us to analyze the D32 allele, which is of European origin (Faure and Royer-Carenzi, 2008), in a population of Quilombolan descendents, although without investigating the relationship between CCR5D32 and AIDS. Notwithstanding, future studies in this region should thus address this issue, given the high incidence of HIV infection.
In conclusion, the European heritage of the individuals comprising the current population of Rondônia was reflected in our observation of the CCR5D32 allele in the Quilombolan population of the Guapore Valley, the ethnically heterogeneous mothers from the Hospital de Base of Porto Velho, and individuals from Candelária and Bate-Estaca, where the current population is descended from Negroid immigrants and native Amerindians.